Holographic interferometry offers tremendous potential in the analysis of small amplitude vibrations of three dimensional objects. Due to stability requirements and the necessity of a coherent light source, the analysis is still limited to the laboratory. In this experiment the various holography systems for vibration analysis developed to date were investigated. Secondly, a real time holographic interferometry technique was developed for quantitatively mapping the complete dynamic field of a surface, where the amplitudes of vibration are on the order of the wavelength of light used. This technique offers the following features: (a) The phase of the object can be determined. (b) The exact resonant frequency can be pinpointed by observing the phase change of the vibrating object (c) Entire displacement fields can be determined, even though the maximum amplitudes of vibrations are on the order of only a few microinches. (d) Under continuous illumination the mode shape of the object can be readily observed. / Applied Science, Faculty of / Mechanical Engineering, Department of / Graduate

Coupling of the flexural, longitudinal, and torsional vibration modes of a column subjected to periodic axial loading was analytically and experimentally investigated. The initial crookedness of the column and longitudinal inertia give rise to coupled flexural-longitudinal vibrations. Further, the Weber effect and longitudinal inertia result in coupling between longitudinal and torsional oscillations. To assess the validity of the theory, an experimental apparatus was set up to axially excite a column using a vibration control generator and an electromagnetic shaker. The experimental results were in good agreement with the theoretical predictions. Coupled longitudinal vibrations exhibiting a frequency ratio of 1:2 were observed. Coupled flexural oscillations were also observed, though a frequency ratio was not established. Further, the experimental results suggest that coupled vibrations other than those theoretically expected were present. In particular, a longitudinal coupled vibration with a frequency ratio of 1:3 was observed, and a corresponding coupled flexural oscillation was present. A coupled torsional mode was experimentally observed when the applied frequency was twice the fundamental torsional frequency. A second coupled torsional mode appeared when the excitation frequency was three times the fundamental torsional frequency. The phase relationship between the coupled vibrations was observed. The resonant coupled vibrations were found to be significant at certain frequencies. / Applied Science, Faculty of / Mechanical Engineering, Department of / Graduate

Floor vibration is a serviceability limit state problem as opposed to limit state of collapse. People find excessive vibrations annoying and it leads to unacceptable floor performance. Improvements in design techniques and use of higher strength materials have made floors lighter and more sensitive to vibrations. The existing code" criteria for limiting floor deflection are not satisfactory as they do not relate to human perception of vibrations. A finite element computer program is developed for a rational analysis of floor vibrations due to human footfall impact. The approach followed in the solution is to use a Fourier series expansion of displacement functions along the span and a finite element approximation in the transverse direction. A time step analysis is followed to obtain the floor's dynamic response at any point on the floor due to footfall impact elsewhere. A sensitivity analysis is done to study the effect of various floor parameters on floor performance. A reliability study of floor performance as affected by variability of joist stiffness is made by simulation. Floors simulated by randomly selecting joists from a population are analyzed and vibration response ratings are obtained by relating the response to human perception criteria. The probability of people rating the floor as of a certain performance level is predicted. Based on the reliability analysis, allowable spans and allowable deflections for a certain level of floor performance and a target reliability are proposed. / Applied Science, Faculty of / Civil Engineering, Department of / Graduate

Analytical and experimental studies of the behaviour of equipment vibration isolators under seismic conditions are presented. A preliminary parametric study of the effect of equipment-structure interaction on the ultimate equipment response of general equipment-structure systems is considered first. The results of this study indicate the conditions under which a non-interactive approach can yield adequate ultimate equipment response estimates. A model of a prototype air handling unit mounted on vibration isolators was constructed for use in the experimental studies. Two types of vibration isolators -elastomeric isolators and open spring isolators with uni-directional restraint - were tested under static and dynamic conditions. The frequency and seismic response characteristics of these vibration isolated systems were obtained. The experimental results indicate that the vibration isolators have nonlinear stiffness characteristics and high damping values. The results also show that the elastomeric isolators can survive a substantially higher level of base excitation than the open spring isolators with uni-directional restraint Analytical models of the vibration isolated systems, based on the model identification test results, have been formulated. A numerical procedure, utilizing time series analysis, was used to solve the equations of motion of the systems. Good agreement between the experimental results and the analytical results was observed. This study indicates that the analytical procedure can be used to accurately predict the response characteristics of vibration isolated equipment systems subjected to known base excitation inputs. / Applied Science, Faculty of / Civil Engineering, Department of / Graduate

No description available.

No description available.

This report tests a new procedure to identify the eigenvalue problem matrix "A" by utilizing the frequency response of a system in state-space for a given sinusoidal forcing input. Two sample identification tests are performed and a detailed explanation of both cases is presented. The method proved to be correct and the results obtained were accurate in both cases. A random error was then added to both sample tests in order to simulate real measuring conditions. Results indicate that it is more difficult to identify the "A" matrix of an undamped system. The error in the system must be very small for this type of system to be identified. However, when the identification is repeated several times and then averaged, the identification becomes more accurate.

No description available.

No description available.

Thesis (Ph. D.)--West Virginia University, 2002. / Title from document title page. Document formatted into pages; contains xiii, 358 p. : ill. (some col.). Vita. Includes abstract. Includes bibliographical references (p. 248-257).